Will Buffet electrify the BNSF? Part III—operation advantages of electric power

(part 3 in a series)

Why use electricity? Rail transport is already very efficient (you've seen the ads)—436 ton-miles per gallon. (FWIW, the average car gets about 40 ton-miles per gallon, trucks do somewhat better.) So, that's good, right? Yes. It's good. But, in addition to easing operation when built, freight rail could triple that number. One ton across the country, on two gallons of gas.

Railroads are already efficient—significantly more efficient than their chief competition: trucks. Pipelines and barge traffic are also quite efficient but each have significant limitations. Pipelines are expensive to construct and can only carry liquids. Barges are cheap to operate and energy efficient (especially going downstream, where they use the flow of a river to their advantage) but are tied to navigable rivers and stream flows, which, when low or icy, can preclude their use. In addition, barges have a very limited top speed, and also need long periods of time to navigate locks when making any change in elevation. Thus, barges are only useful for bulk materials which are not time-sensitive. To receive or deliver goods anywhere which is not on the barge network requires time-consumptive and expensive break-in-bulk procedures, which, when combined with the restrictiveness of the navigable waterway network, further decreases their utility.

So, highways and railroads handle the bulk of freight in the United States. Trucks have advantages in flexibility (they can deliver almost anywhere) and, generally, speed. Railroads have advantages in fuel consumption, labor costs, and maximum carrying capacity (by unit; the size of the largest rail car is significantly more than a trailer). While labor costs and maximum capacity would be relatively unaffected by electrification, fuel costs would decrease further. Using diesel power, railroads are already between 1.5 and 10 times more efficient than trucks. A factor of three or four is probably a conservative estimate. Railroads and trucks use the same fuel, so the efficiencies are not realized there. They appear in both economies of scale of railroads larger engines, wind resistance (in effect, each rail car is drafting the one behind it) and, more importantly, the effect of rolling resistance. Rubber tires on asphalt roads have significantly more friction than steel wheels on steel rails.

Even with these efficiencies, railroads are generally cheaper than trucking because of labor costs. Each truck requires a driver, and a train, which can carry the equivalent of 280 trucks with a crew of two. With current energy prices, labor is a greater advantage for railroads than fuel. But it doesn't mean that diesel power is operationally preferable to electricity. Once the initial infrastructure (catenary, transmission and substations) is built, electric rail is operationally superior for several reasons including the simplicity of electric motors, the lack of a need to ship fuel, acceleration and operating speeds, and, finally, the ability to use regenerative breaking on downhills.

The first reason is that electric motors (technically, electric train engines are "motors") are simple. As discussed in part II of this series, many of the electric motors the Milwaukee Road used were fifty years old and worked fine when the railroad ripped out electrification. Diesel engines last rather well, too, but aren't in the same league. With fewer moving parts, after the initial investment, a railroad could expect to have to pay very little for new motors for some time.

A second advantage is where the power for engines comes from. With diesel engines, there is both the need to carry fuel on-board, and to frequently refuel. The weight of the fuel on the train itself is quite minor, considering a train might weight several thousand tons. However, the transport of the fuel requires resources, either pipelines or delivery by the railroad, which uses capacity that could be used for other shipments. In addition, fueling the tanks takes time, during which the engines could otherwise be in service. With electricity delivered from overhead wires, there's really no reason, except for crew changes, that a motor would ever have to stop.

Furthermore, when diesels do have to stop, they can't be turned off and back on at the drop of a hat. Diesel requires warm temperatures to operate, and to keep engines warm, they either have to be plugged in or kept running, whether they are hauling anything or not. Electricity, on the other hand, is as easy as flipping a switch. In the mountains and along the northern transcontinental route, it gets mighty chilly.

Electric motors benefit from better acceleration and higher operating speeds. Acceleration is very important for passenger rail, especially when there is not much distance between stops (which is why subways run on electricity) but not as important for freight rail. However, having a top speed faster than competing services would allow freight rail to be time-competitive with trucking.

Getting rid of the on-board power supply also gives the ability to reduce the dependence on one fuel type, which, in the case of rail, is oil. Diesel-electrics use on-board power plants, which are only 30 or 40 percent efficient. Some electricity-generating technologies are more efficient. (HowStuffWorks has a nice article on diesel locomotives.) The cynic's view is that railroads will turn to coal in order to get their power, and, while this may be true (they're the ones hauling the coal, after all), there are certainly other options. The Milwaukee Road ran mainly on hyrdo power. As we'll explore later, the BNSF runs through wind- and solar-heavy regions. Finally, since there is some power loss, having major, centralized coal plants might not make as much sense as power sources along the route.

Finally, a diesel engine runs whether the train is accelerating or not. If the train is decelerating, the engine can, in a sense, be run backwards to slow it down. (Your car runs very similarly, albeit on a smaller scale.) This is called "dynamic braking." Of course, physics dictates that this energy has to go somewhere, and it does: it is converted to heat and blown through huge vents on the top of the engine. This is a major waste of energy.

Electric engines also have the ability to use the momentum of the train to slow it down, but instead of dissipating the energy as heat, they throw it right back in to the wire above. This is "regenerative braking." (The Prius does the same type of thing, but can only store energy in a battery.) With wires above, there is little limit to the amount of energy which can be put in to the system. If another train on an adjacent track is climbing the hill, a downhill train can transfer much of its power across to it; if not, the power can be fed back in to the grid. Since every transcontinental line climbs and descends several thousand feet through the Rockies and coastal ranges, there is the potential to save huge amounts of power.


I. We've discussed how being part of a larger organization like Berkshire Hathaway may allow the BNSF to spend more freely on capital improvements in this section.
II. We'll then look at a history of freight rail electrification, including the sad tale of the Milwaukee Road and some freight rail electrification abroad.
III. We'll look at some of the operational advantages of electric power, and
IV. Some of the economic advantages, in the long run, of electric power generation, and how the whole system would be built.
V. From an environmental standpoint, we'll look in to how electricity can be generated on-route, and whether there are options beyond coal (such as wind and solar), and
VI. How this may mesh with the construction of a smart grid.
VII. Finally, we'll see if freight rail electrification may have any benefits for passenger rail, on the BNSF routes and other main lines.